Closed Jeff-Thompson12 closed 1 year ago
Can now make a table like this:
"test_1": {
"title": "Test 1: Use ATPT",
"blocks": [{"data":"ADVS", "variable":"DIABP", "var_arg":"ALL", "var_options_fn":"atpt", "statistic":"MEAN", "stat_arg":"ALL", "stat_options_fn":"avisit"}]
}
I am converting this back to draft because I am realizing I need to make sure that the user can specify var_options
and stat_options
in the recipes JSON file.
All right. I think I have this working now. If you add a recipe like this:
"test_2": {
"title": "Test 2: Use selection and options",
"blocks": [{"data":"ADVS", "variable":"DIABP", "var_arg":"AFTER LYING DOWN FOR 5 MINUTES", "var_options":{"ATPT":["AFTER LYING DOWN FOR 5 MINUTES"]}, "statistic":"MEAN", "stat_arg":"Week 4", "stat_options":["Week 2", "Week 4", "Week 6"]}]
}
You will get the appropriate outputs in the table generator.
@AARON-CLARK Did you get a chance to look at this PR yet?
(2) Validation. I know it would probably be some work, but can we check to make sure these metadata inputs make sense? I changed the
var_options
from your example to "7 minutes" and left everything else the same and the output came out like this.
Big problem with validation is that the block data is created separate from the dataset. For this specific case I can add in a check that the var_selection
is contained in the var_options
. There is the question of where this check should take place since the user could supply var_options_fn
instead. Currently the output does suggest that something went wrong and maybe even suggests what that was:
I do agree that instead of expanding upon the recipe metadata that the next step should be validation of what has already been built out though.
Also, did I do something wrong? I use your example
yml
that I edited with the new naming convention:
No, that's on me. I will fix it.
Codecov Report
0.00% <0.00%> (ø)
0.00% <0.00%> (ø)
0.00% <0.00%> (ø)
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.